Unexpected severe incident vertebral fractures in patients with systemic lupus erythematosus: comment on the article by Zhu et al. Sir, We read with great interest the article by Zhu and colleagues about the incidence of and risk fac-tors for clinical non-vertebral and radiological ver-tebral fractures (VFs) in Chinese women with systemic lupus erythematosus (SLE) published recently in Lupus. 1 We would like to compliment the authors with their prospective cohort study on this important subject. Earlier prospective studies were focused on incident symptomatic fractures in SLE. However, studies focusing on symptomatic VFs have a disadvantage in that only one-third of all VFs come to clinical attention, 2 which may lead to an underestimation of VF occurrence in patients with SLE. Identifying vertebral deformities is important since VFs are associated with reduced quality of life in postmenopausal women, 3 and an increased mortality rate and increased risk of future (vertebral and non-vertebral) fractures in the gen-eral population. 4,5 Recent cross-sectional studies have demon-strated a high prevalence (20–50%) of VFs in SLE. 6–8 In contrast to the high prevalence of radio-logical VFs found in lupus patients in cross-sectional studies, Zhu and colleagues report a very low incidence of radiological VFs (0.94 per 100 patient-years) in their study cohort. However, the authors do not give a clear explanation for this finding. Moreover, the authors report that the majority (4 out of 7) of the incident VFs occurring in their study patients were severe fractures, defined as >40% reduction in vertebral body height (Grade III), which is a very unusual finding since earlier studies on VFs in SLE using the same semi-quantitative scoring method for scoring vertebral deformities demonstrated the majority (73–79%) of VFs to be mild fractures (Grade I, 20–25% reduction in height). 6,8,9 The authors do not discuss this remarkable finding in their paper. While they report that all but one of these incident VFs were asymptomatic, the severity of these fractures sug-gests a traumatic cause. Zhu and colleagues report that higher organ damage and lumbar spine osteoporosis are risk fac-tors for incident VFs in their study. However, the association found between higher organ damage index, as assessed with the SLICC ACR damage index (SDI), and the occurrence of incident VFs could be considered as a circular argument since the SDI comprises osteoporotic fractures as a damage item. Therefore, the authors should have used a modified SDI score excluding osteoporotic fractures as a damage item. Despite these comments, the results of the study by Zhu and colleagues illustrate the importance of further research on the incidence of and risk factors for VFs in larger groups of patients with SLE, since these fractures are often asymptomatic but clinic-ally important in terms of morbidity, mortality, and future fracture risk. Moreover, the detection of one or more asymptomatic VF in patients with a bone mineral density in the osteopenic range might be an indication for starting treatment with anti-osteoporotic drugs to prevent future fractures.